1 Freedom and Strength of Will: Descartes and Albritton Paul Hoffman In his intriguing and entertaining APA presidential address, Rogers Albritton defends the Cartesian view that the will is so free in its nature that it cannot be constrained. But Albritton's account of freedom of the will differs in some fundamental ways from Descartes's account. This afternoon I shall examine the differences between the two accounts. In defending absolute freedom of the will, Albritton distinguishes freedom of will from two other notions with which it is often confused. First, he argues decisively against the pervasive tendency to identify freedom of will with freedom of (bodily) action. To restrict someone's freedom of movement, by putting him in chains or even by administering curare, is not to diminish the will's freedom. Such measures diminish only the efficacy of the will, that is, they disable the will, but they do not affect its freedom. On Albritton's view, our will is free if what we propose to do is up to us. Our will is efficacious if our proposing to do something brings about its being done. Albritton is surely correct that we might freely propose to do something that does not get done because our will is ineffective. Second, Albritton distinguishes between freedom of will and strength of will. He invokes this second distinction to block the claim that lacking the strength of will to resist doing something is a case of unfreedom of the will, but he does not attempt to justify this second distinction. In defense of Albritton, one might well be tempted to link the will's strength with its efficacy. A weak will is presumably less effective than a strong will. If someone has a weak will, what he proposes often does not get done. Although a weak will need not be completely ineffective, the things that a weak-willed person proposes that do get done must be relatively easy. A person with a strong will, on the other hand, proposes difficult things, and what he proposes gets done. Not even a strong-willed person can be expected to overcome every obstacle, so perhaps his ultimate goals are not achieved. But in that case he must accomplish some things that count as working toward those ultimate goals, and to that extent his will must be efficacious. In so linking strength of will to efficacy of will, these considerations suggest that a weak will might still be fully free. What I propose to do might be fully up to me, even if I am not very effective in carrying out that proposal. On this understanding, strength of will has to do with the output side of will. What I mean by this can be illustrated in terms of Descartes's conception of the will. He says that our volitions come directly from our soul (I 17), so in some sense the soul is their source. But he distinguishes between volitions that also terminate in the soul and those that terminate in the body (I 18). Volitions that terminate in the body are those we normally associate with human action, where this is understood by Descartes as falling into two broad categories: pursuit and avoidance. What I mean by the output side of will, then, is the bodily action in which a volition terminates. One might think of volitions that terminate in the body as being in competition with other internal sources of bodily action. These other sources of bodily action might be conceived, as they were by Plato and Aristotle, as other parts of the soul. Or they might be conceived, as they were by Descartes, as something in the brain (I 47). A strong will is one that tends to win these struggles; a weak will is one that tends to lose them. So, for
2 example, a volition that I stand my ground in the face of some danger might be in competition with an effort by some other part of me to make me flee. That I do stand my ground is an indication that my will is strong; that I don't is an indication that my will is weak. When one thinks of the will as being in competition with other internal sources of bodily action, it is clear that its efficacy depends on its strength on the output side. There is, however, a contrasting conception of strength of will according to which it has to do not with the output side but with the input side. So we often think of a strong-willed person as someone who resists outside forces in making decisions or in sticking to them. These outside forces need not be outside the person. They might be conceived as being inside the person but outside some privileged part or aspect of the person. Thus a strongwilled person is often regarded as someone who listens to the voice of reason, understood as a part or aspect of the self, while resisting the passions. So long as strength of will is considered from the output side, I agree with Albritton that freedom of will is independent of strength of will. But it is far from clear that freedom of will is independent of strength of will considered from the input side. According to Albritton, our will is free, if what we propose to do is up to us. It is not obviously wrong to say that if our will is weak on the input side, what we propose to do isn't really up to us. What we decide to do is instead determined by an external force. Indeed, when Descartes says that certain remedies against excessive passions prevent the soul from losing its free judgment, he implies that without such remedies the passions or at least their excesses could cause us to lose our free judgment (AT IV 411; CSM-K 287). Even though Descartes and Albritton both claim that our will is so free in its nature that it cannot be constrained, they thus apparently differ on the relation of the will's freedom to its strength. Albritton believes on conceptual grounds that weakness of will is no barrier to freedom of will. Descartes, however, believes that one must take certain practical steps to insure freedom of will. Underlying this disagreement is a fundamental difference in their conceptions of free will. Let us begin with Descartes's conception. Descartes defines the will not only as the power to pursue or shun but also as the power to affirm or deny (AT VII 57; CSM II 40). Thus the will has to do with judging or believing (affirming or denying) and with choice (pursuing or shunning). He thinks that the will is inclined to affirm what appears true, to deny what appears false, to pursue what appears good, and to shun what appears evil. The more clear and distinct the perception, the stronger the inclination. So Descartes asserts that our will cannot tend toward anything else but truth and goodness and that a person embraces what is true and good more willingly and freely in proportion to seeing it more clearly. He asserts that we are at our freest when a clear perception impels us to pursue some object (AT VII ; CSM II 292). Finally, he asserts that "the will of a thinking thing is drawn voluntarily or freely (for that is the essence of will), but nevertheless infallibly, towards a clearly known good" (AT VII 166; CSM II 117). Descartes maintains that our sensory ideas þ our sensations of color, sound, heat, cold, and the like þ are obscure ideas that, by a natural impulse, prompt us to make judgments that things resembling them exist in bodies external to us (AT VII 38; CSM II 26). These judgments are false, but he thinks we can refrain from making them because it is within our power to withhold our assent from ideas that are obscure or confused (AT VII 59; CSM II 41).
3 Our passions are like our sensations in that they are obscure or confused perceptions (I 28). Descartes sometimes talks as if they have an influence on the will in virtue of being representations of things as good or as evil (II 138). An idea that represents something as good might influence the will in two different ways. First, it might influence us to make the judgment that the thing is good. Second, an idea that represents something to us as good might influence us to pursue that thing. Suppose that we have made the judgment that something is bad, but now a particular passion represents it to us as good. Descartes thinks we might pursue that thing. That is, we might choose in accordance with our present passion instead of choosing in accordance with our judgment. This would seemingly be an instance of weakness of will. Indeed, Descartes defines strength and weakness of souls in terms of their ability to follow firm and decisive judgments concerning good and evil and to resist present passions opposed to those judgments (I 49). To the extent that present passions are seen as alien forces operating through the will, it seems plausible to say, as Descartes does, that a will that obeys them is enslaved (I 48). It seems plausible even if we suppose that the will could have resisted them successfully but failed to. The failure to resist the passions successfully in choosing what to do can be seen as analogous to other sorts of failures in which it seems wrong to say that the outcome is something I do freely. If a long jumper lands short of a distance which, on the basis of his having reached it on his previous attempt, we reasonably suppose he could have reached, it seems wrong to say that that fact by itself shows that he landed short freely. Or if I don't succeed in my attempt to suppress a sneeze, even if I might have succeeded, it still seems wrong to say that my sneezing is something I do freely. By the same token, I don't will freely to run from a burning building if I do so out of fear, even if I overcame that fear yesterday in a similar situation and remained long enough to carry someone to safety. To be free and not enslaved, a weak will therefore needs a way of transforming the passions so that they are no longer alien forces. Descartes claims that by using techniques most closely associated these days with behavioral psychology, even the weakest soul can gain control over which passions it has (I 50). We þ that is, our bodies or the connections between our souls and our bodies þ can be "rewired" in such a way that a given first cause produces a passion different from the one it would naturally produce. It is in virtue of this capacity to determine which passions they have that souls are claimed to have absolute control over the passions. Any force over which we have absolute control is ours; it is no longer alien to us. Albritton, although he does not say so explicitly, rejects the Cartesian view that the will is inclined only to those things that have some appearance of goodness (I 177) and that the will is drawn infallibly toward a clearly known good. He seems to think that if the will were inclined only to things of a certain sort or infallibly drawn toward anything, what we propose wouldn't be up to us, and in that case the will wouldn't be free. This feature of Albritton's view is revealed most clearly in his response to the objection that "every decisive, compelling reason to make one choice rather than another reduces one's freedom of will." In complete contrast to Descartes, who holds that being compelled by reasons, in the form of clear and distinct ideas, is the highest form of freedom, Albritton holds that it is always open to us to decide not to do what reason or anything else dictates. Albritton thus seems to be identifying freedom of will with what is known as the liberty of indifference. Liberty of indifference is typically defined in terms of acting: a free agent is one who, all the conditions required for action having been posited, can either act or not, or perform a contrary action. But liberty of indifference could easily be redefined in terms
4 of choosing: our will is free if, all the conditions required for choosing having been posited, we can either choose or not, or choose the contrary. Underlying these rival conceptions of freedom þ Descartes's compulsion account on the one hand and the liberty of indifference on the other þ are, it seems to me, different conceptions of the self. Descartes believes that reason or intellect is our essence. So to be compelled by reason, in the form of its clear and distinct ideas, is to be compelled by oneself. But Albritton, in his denial that we are compelled by reason, seems to conceive of being compelled by reason as being compelled by something external to us. Descartes thus has an internalist account of reason þ it is internal to the self þ whereas Albritton has an externalist account. Albritton, however, does not tell us what he thinks the self is, so not only is there no positive theory on which to evaluate his view that reason is external to the self, but it also is not clear what he means by saying that what we propose to do is up to us if our wills are free. Up to what exactly? In Descartes's case, at least we know what this means: up to reason or intellect. Of course, one's sympathies might well be with Albritton here. Even if one does not go so far as to reject as misguided any attempt to provide a positive ontological theory of the self, one might agree, as I do, that it is a mistake to reify reason and incorporate it into the self. But even if one went so far as to suppose that we do not need any positive account of the self to understand what it is for something we propose to be up to us, one fundamental problem with the liberty of indifference is that it posits "a requirement of indetermination" that seems to detach what is proposed from the rest of us in such a way that it does not seem plausible to say that it is proposed by us. This problem would seem especially acute with respect to beliefs and desires. How could what we propose to do be up to us if it is not in accordance with our beliefs and desires? Indeed, it would seem that if what we propose to do is really up to us, not only must it be in accordance with our beliefs and desires, but it must be caused by our beliefs and desires. One strategy to reconcile the liberty of indifference with the thesis that our choices are caused by our beliefs and desires is to deny that being caused to do something entails being compelled by it. So one might assert that our beliefs and desires can cause us to propose one thing rather than another and deny that they can compel us to propose that thing. One might also maintain that reasons can be causes of what we propose, while at the same time denying, as Albritton does, that they compel us. By holding that we are free so long as we are not compelled, one could argue that we are free even though what we will is caused by our beliefs (or by reason). Albritton seems to acknowledge that there is a tight connection between belief and will when he says, "One can leave the will alone and get excellent results even now, by manipulating belief instead. Convince me that your enemy is the Antichrist and I will no doubt behave satisfactorily, in full freedom of will. How else should one behave toward Antichrist?" It is not clear, however, how much Albritton is conceding here. In particular, it is not clear that he is granting that our beliefs can cause our choices. First, there are indications that he is uncomfortable with the compatibilist attempt to draw a distinction between what we are caused to do and what we are compelled to do. So he says that he is foggy about the idea that a free decision might nevertheless be determined by natural laws, "possibly because I am bogged down in superstitions about natural law and the causal nexus." But perhaps this remark does not reflect a concern about the coherence of supposing that a free decision is caused by other events or states, such as beliefs, but instead reflects a concern about the coherence of supposing that a free decision
5 is determined by natural laws. In other words, it reflects a concern about a specific kind of causation: determination by natural laws. Second, Albritton maintains that the connection between our choices and one important class of beliefs is not causal but conceptual. Our beliefs about what is impossible can constrict our choices without interfering with our freedom of will because, Albritton claims, it is a conceptual truth that we cannot choose contrary to such beliefs. He says, `it seems to me not deficiency of free will that one can't just up and go against knowledge and belief (insofar as one can't) because that "can't" is again, not psychological or metaphysical either, but "grammatical" or "conceptual".' Albritton claims not only that it is a conceptual truth that we can't decide to do what we believe is impossible, but he also claims that what we can try to do is conceptually constrained by our beliefs: "Trying to walk isn't a perfectly imaginable little act of the will, separate and distinct from all belief and contingently blocked in him by belief." There is reason to attribute to Albritton the view that the connection between our belief that someone is the Antichrist and our subsequent decision to behave toward him in a certain way is also conceptual and not causal. Otherwise it is hard to see on what grounds he can maintain both that manipulating us to have that belief will produce excellent results and that it will not interfere with our freedom of will. However, this example is especially troubling for our understanding of Albritton, because the belief that someone is the Antichrist seems to influence our decision to behave toward him in a certain way in virtue of providing a reason (or at least connecting us to a reason) for behaving toward him in that way. If our choices are conceptually connected to our beliefs in virtue of those beliefs being connected to reasons for those choices, then it would seem to follow that those reasons are also conceptually connected to those choices and hence that those reasons determine our choices. But Albritton denies that reasons determine our choices. On the other hand, to say that manipulating our beliefs provides excellent results is not necessarily to say that it provides infallible or inevitable results. So perhaps Albritton's view is that manipulating beliefs provides the results it does not because the connection between those beliefs and our choices is conceptual but instead because it is a fact about human beings that most of us usually act in accordance with overwhelming reason of the sort that we have when we believe that someone is the Antichrist. But this fact, if it is a fact, cries out for explanation. Why do most of us usually choose in accordance with reason? This question points to still a further problem with the liberty of indifference. Understanding liberty as indifference precludes an important kind of explanation of human behavior. It does not preclude all explanation of human behavior - when someone does choose in accordance with reason, we could still explain his behavior by appeal to those reasons. But if freedom requires indifference, if it requires that we can always choose contrary to reason, then the fundamental question of why someone does or does not choose in accordance with reason is unanswerable. Albritton seems implicitly to concede this when he says: How very odd that the obvious reasons to stop behaving in this way don't weigh with him as decisively as one might expect! Well, yes, it is odd, but there it is: they don't, and he doesn't stop... But having to do a thing does not settle magically the question whether to do it or not. Reasons, of whatever species, logically can't close that question. It's a question of a different genre, and is not relative to any system of reasons. It isn't for reasons, in the end, that we act for reasons.
6 One's philosophical sympathies on this point might once again lie with Albritton. It might be seen as a virtue of his theory that we should cease looking for an explanation of why we do or do not choose in accordance with reason. But I don't think freedom of will should require us to be such mysterious creatures as that. It is interesting that in defending a more radical view of free will than Descartes, Albritton allows for manipulation of belief of the sort that Descartes, at least according to his official view in the Meditations, would reject as impossible. Albritton distinguishes believing from willing, whereas Descartes, as we have seen, identifies believing with judging - that is, with assenting and denying, which are modes of willing. On Descartes's theory, to manipulate our beliefs would be to manipulate our will. But since he thinks that all clear and distinct ideas are true and that we can always refrain from assenting to an idea that is not clear and distinct, he is able to maintain that all false beliefs are ultimately "up to us." They are our own responsibility. Albritton's treatment of the relation between desire and choice differs drastically from his treatment of the relation between belief and choice. On the one hand, he claims that to the extent desires are taken to give us reason to behave one way or another, they are as powerless as any other species of reason. On the other hand, he compares acting from desire or acting out of fear with being put in chains or being violently thrown into bed in order to suggest that in such cases there is no interference with freedom of will because we haven't done anything. That is, he is inviting us to suppose that when we act from desire or out of fear, choice is bypassed. By holding that choice is bypassed when we act out of passion, Albritton is in a position to claim that the passions can interfere only with our freedom of action, not with our freedom of will. Indeed, Gary Watson claims to have learned from Albritton that it is a mistake to think of one's own desires and emotions as potential impediments to free will, because "however internal, these obstacles are still obstacles in virtue of their (potentially) getting in the way of implementing one's will." For our purposes, this is perhaps the most fundamental difference between Descartes and Albritton. Descartes thinks that present passions influence behavior by operating through the will. They influence us in making choices, they don't affect merely the implementation of those choices. Since the passions operate through choice, they are potential impediments to free will. To be sure, when there is a conflict between our firm and decisive judgments concerning good and evil and our present passions over which will govern our choices, those passions are potential impediments to the implementation of those judgments, and therefore, since Descartes also considers judgments to be acts of will, it follows that he thinks the passions are potential impediments to the implementation of our will. This might make it seem as if the dispute between Albritton and Descartes over whether the passions bypass the will or operate through the will is merely terminological. If we can construe Descartes in speaking of judgments concerning good and evil to be referring to the same thing as Albritton in speaking of proposals to do something, then it looks as if they are in basic agreement that passions are obstacles to freedom in virtue of interfering with the implementation of our will. But the differences between Descartes's and Albritton's accounts of the relation between the passions and the will are not merely terminological. First, judging that something is good is not the same as proposing or choosing to do it. It is a mistake to think of Cartesian judgments as choices. Second, and here lies the crux of the matter, what underlies the dispute between Albritton and Descartes over whether the passions operate through choice
7 or bypass choice is a dispute over whether we are doing something when we act out of passion. Descartes thinks we are, Albritton thinks we are not. Third, Descartes thinks that the passions can influence us in making judgments as well as in making choices. Thus they can affect the input side of judgment as well as the output side. To allow as Descartes does that passions influence us in making choices and in making judgments opens the door to the possibility that they can interfere with our freedom of will. Descartes's assertion, mentioned earlier, that a soul that obeys its present passions is enslaved does provide important evidence that he thinks weak souls which have not made use of his behavior modification techniques are unfree. But that is only one of several characterizations he gives of the relation between the passions and the will, which are in turn subject to various interpretations as to their implications for the freedom of weak souls. Sometimes Descartes suggests that we should think of the passions as forces acting on the will. For example, he says that the passions "almost always make both the goods and the evils they represent appear much greater and more important than they are, so that they incite us to seek the former and flee the latter with more ardor and more anxiety than is suitable" (I 138). This language strongly suggests, although it certainly does not entail, that Descartes thinks the passions can cause our volitions to pursue or to shun. An additional reason for ascribing to Descartes the view that the passions can cause our volitions is provided by the analogy of passions with sensations (indeed, he defines passions as sensations, I 27-29). In the Third Meditation he describes our judgments based on sensations as resulting from natural impulses that push us in certain directions (AT VII 38-39; CSM II 26-27). That language certainly sounds causal. Thus if the passions' relation to the will is similar to that of other sensations, it too would be causal. If our passions cause our volitions, and if those passions are in turn caused, as Descartes thinks most of them are, by things external to us, then I believe Descartes would argue, and I would agree with him, that the resulting volitions are not up to us and hence are not free. I believe that this conclusion follows even if we suppose, as I believe Descartes does, that the notion of being caused is distinct from and weaker than the notion of being compelled. If our will is free only if what we choose to do is up to us, then freedom of will would seem to require not merely that we are not compelled to make a given choice by something external to us but also that we are not caused to make a given choice by something external to us. Descartes does claim that some of our passions have their origin in our decisions to think about this or that (I 51), and it would seem that any volitions caused by those passions would be up to us. If, moreover, we make use of the behavior modification techniques to rewire ourselves, then the passions resulting from that rewiring and the volitions they cause would be up to us and hence free, even if our soul remains weak because our volitions to pursue or to shun are not caused by our firm and decisive judgments concerning good and evil. Descartes's assertion that the soul will be enslaved if it obeys present passions invites us to think of the passions not as forces acting on the soul but as potential masters to be obeyed or disobeyed. To think of the passions as the master and the will as the slave is different from thinking of the passions as causes of acts of will. In commanding the slave to behave in a certain way, the master does not cause the slave's behavior. It remains true nevertheless that the slave is not free. What the slave proposes to do in obeying the master is up to the master rather than the slave. Even though it remains in the slave's power to
8 disobey, it would be perverse to draw the conclusion that his decision to obey is free. He is not free to disobey because of the sanctions attached to disobedience. To say that the slave is not free is not necessarily to imply that the slave does not have free will. Suppose the slave disobeys (as a matter of principle, not just out of anger). In that case what he does is up to him, so it seems right to say that he disobeys freely. But it still seems right that he does something he is not free to do (even if we agree that it was within his rights to disobey). He is not free to disobey, because as soon as he does, he is going to be constrained by the master. But if the slave were sufficiently strong that he could not be constrained, he would be free to disobey. So the slave's freedom or lack of it depends on his strength or lack of it. If he can overcome the master, he is free; if not, he is not. Now if our passions could constrain a disobedient will in something like the way a master constrains a disobedient slave, then it seems right that our will is no longer free to disobey them. The will has the power to choose to disobey, but if it does, it will be thwarted. If this is right, then it follows, contrary to my original agreement with Albritton, that freedom of will does depend on the will's strength on the output side, in the same way that a slave's freedom is blocked by a lack of strength. Moreover, we can't intelligibly say of the will, as we might of the slave, that even though it isn't free, it still has free will. But we might say of the will that whether it disobeys is up to it. So even if the will is not free to disobey the passions, it might still disobey them freely. Perhaps the lesson here is that there are two notions of freedom of will, one corresponding to strength on the input side and one to strength on the output side. The slave analogy shows how a will might be unfree because it is not sufficiently strong on the output side. To see how a will might be unfree on the input side, we need to rely on one of Descartes's other characterizations of the relation between the passions and the will þ for example, that the passions are a force acting on the will. In any case, if the passions can diminish the will's freedom in the same way that a master diminishes the freedom of the slave, it must be because the passions can apply sanctions to the will should it disobey. At this point we get no help from Descartes. He does not push his metaphorical language so far as to discuss how the passions might respond if disobeyed. He does say that the passions compel us to dwell upon certain thoughts, which could be construed as a particularly harsh kind of sanction, except that he makes that assertion in the context of explaining what the passions are (II 102). Still a third way in which Descartes characterizes the relation between the passions and the will is to say that a weak soul "continually allows itself to be carried away by present passions." This language suggests that the soul either is seduced by the passions or allows itself to be seduced by the passions. Allowing oneself to be seduced or even being seduced is a far different matter from being enslaved, and its implications for loss of freedom are much less straightforward. There is some temptation to say that if the soul allows itself to be carried away, if it allows itself to be seduced, then it is the soul that is the master and not the passions. The passions are a force over which the soul has control, but which it allows to act on it. On such a picture it seems correct to say that what the will proposes is still up to it, even when it follows present passions, and hence that even a weak soul remains free. On the other hand, it is not so clear that being seduced is correctly characterized as something that we allow to happen to us. If we allow something to happen, then we really aren't being seduced. Seduction is another way in which our will is overpowered. When we are seduced, what we propose to do is not really up to us, it is up to our seducer. On this way of looking at seduction, it does imply a loss of freedom. And even if it is sometimes